COVID-19: Modelling UpdateCOVID-19 Rate per 100,000, Last 7 Days (to Nov 18, 2020) In the last seven...
27
COVID-19: Modelling Update November 19, 2020 1
COVID-19: Modelling UpdateCOVID-19 Rate per 100,000, Last 7 Days (to Nov 18, 2020) In the last seven days: • 1,231 cases (+645) • 14.5/100,000 avg new case rate • 25 new hospitalizations
• Modelling cannot predict what will happen. • It tells us what could happen over the next six
months, based on actions taken.• It supports decision makers by comparing the
potential impact of public health measures and behaviour changes.
• Goal: to make the best policy decisions based on public health information, keeping economic activity stable, children in school, and not exceeding the health system’s capacity to care for both COVID-19 and non-COVID-19 patients.
Presenter
Presentation Notes
Modelling is not a crystal ball – it is a tool we use to plan our response to the pandemic event. By projecting different scenarios, modelling helps us to figure out how to slow down pandemic transmission based on the patterns and the public health measures that are in place currently and how additional measures may have an impact. Transmission patterns are a complex web of interactions including biology of humans and virus, the environment (i.e. settings in which transmission is occurring), human behavior, and the determinants of health. At this time, there is no evidence of population herd immunity so we must rely on public health measures to prevent spread of COVID-19. If the population does not take the necessary steps to prevent transmission, all models indicate exponential growth over time. This would ultimately overwhelm our health system.
Assessing our Situation: COVID- 19 Indicators
3
Presenter
Presentation Notes
Forecasts must rely on the most recent information available, even though the most recent statistics are as a result of actions that occurred 7-10 days ago. We are presenting here on the last seven day reports that are available.
Saskatchewan’s Epidemic CurveMar 11 – Nov 18, 2020 (5,553 cases)
Total
Active
Presenter
Presentation Notes
Cases, hospitalizations and ICU increasing Highest number of cases reported in one day ever - 308 reported on Sat Nov 11 Highest 7 day daily average ever of 176 on Wed Nov 18 Highest hospitalizations (76) and ICUs (17) on Wed Nov 18 Three new deaths this week - Sunday (2) and Wednesday (1) Exposures for most new cases are unknown or pending The number of active outbreak investigations have increased rapidly
To date:• 5,553 total cases• 2,099 active
cases• 229 in hospital,
67 in ICU• 32 deaths (0.6%
fatality rate)
Source: Public Health Agency of Canada, Nov 18, 2020 https:https://health-infobase.canada.ca/covid-19/epidemiological-summary-covid-19-cases.html?stat=rate&measure=total_last7#a2
COVID-19 Active Cases per 100,000(Nov 18 2020)
Presenter
Presentation Notes
These next two slides are Public Health Agency of Canada slides, available weekly. Saskatchewan has moved to third highest case rate per capita in the last week.
COVID-19 Rate per 100,000, Last 7 Days (to Nov 18, 2020)
In the last seven days:• 1,231 cases
(+645)• 14.5/100,000
avg new case rate
• 25 new hospitalizations8 new ICUs
• 3 deaths
Source: Public Health Agency of Canada, Nov 18, 2020 https:https://health-infobase.canada.ca/covid-19/epidemiological-summary-covid-19-cases.html?stat=rate&measure=total_last7#a2
Presenter
Presentation Notes
Acceleration has been marked in the last seven days.
Area Epidemic Curves, Feb 21– Nov 15
Presenter
Presentation Notes
Provincial trends by the regions. Demonstrates experience of COVID clusters to date – La Loche and communal living settings are spring and summer cases in Far North and South Now is the first wave, looking to North, Saskatoon and Regina (these are representative of the communities where we first implemented mandatory masking).
COVID-19 Hospital, ICU and Deaths by Age Groups (to Nov 18, 2020)
Nov 18 hospitalizations
Total ICU inpatients
19 and under 1 1
20 to 39 6 6
40 to 59 17 7 10
60 to 79 27 9 18
80+ 25 1 24
Total 76 17 59
Presenter
Presentation Notes
While the serious illness and deaths follow the model seen since the beginning of the event -- older and people with co-morbidities – there remains a commonly heald belief that young people do not get seriously ill. Every hospitalization increases pressure on our health care system. We know that there are people with serious, long-term effects of COVID-19 illness, even if they never require hospitalization.
COVID-19 Hospital and ICU Case Census, Aug 1 – Nov 17
Hospitalization
ICU
In-patient hospitalization and ICU counts are trending upward significantly since the beginning of October
Presenter
Presentation Notes
Demonstrates the growing, recent demand that accompanies the surge in cases. Just like it takes 7-10 days for measures to show results in confirmed case trends, it takes longer for those same measures to bring down hospitalization and deaths – another week after that case trend starts to decrease.
COVID-19 Exposures by Type(to Nov 15, 2020)
CasesNov 9-15, 2020
Total Cases to Date
Health Care Workers 16 148International Travel 1 154Domestic Travel 0 112Health Care Setting 0 44Congregational/Communal Living Setting 1 681Mass Gatherings 1 240Workplace 8 193Household Exposure 3 842Social Exposure 6 238Public Facilties 7 175School/Daycare 7 28Unknown 4 285Pending 1,062 2,190
Note: total case totals are more than the reported total of cases as HCWs may be counted in other exposure categories also, depending on exposure.
Presenter
Presentation Notes
NOTE: Household exposure (and communal living settings) versus other locations – private gatherings in the home are a significant source of transmission. These are our close contacts – people we share close quarters with, share meals with, etc. Compared to more controlled settings like acute care settings or schools and daycares.� We are unlikely to see all the “pending” cases assigned due to 1) people having increasing numbers of contacts – we cannot identify the connection between cases – and 2) as contact tracing resources become more strained, there is less capacity to do the required investigations
COVID-19 Cases with Unknown and Pending Exposure (to Nov 15, 2020)
Presenter
Presentation Notes
The increasing trend line: As local transmission increases, and people maintain higher numbers of contacts, it becomes more and more difficult to ever pinpoint the source of transmission. As case numbers increase, contact tracing capacity is consumed – investigations cannot be completed in the timely manner to ensure all contacts are notified, which can result in more transmission.
COVID-19 Outbreaks (Oct 14 – Nov 18)
16 new outbreaks reported week of Nov 9 – 15:- Nine community, four schools, three in other settings
Description (excludes schools and child care) Confirmed OutbreakAssisted Living/Home Care 5
Place of Worship 3Events (Party, Wedding, Funerals) 6
Bingo 1Retail 6Gym 4
Restaurant/Bar 10Sport 8
Presenter
Presentation Notes
Definition of outbreaks – two or more cases, etc. Outbreak reports indicate a high number of adults attending workplaces (restaurants, care settings, schools, etc.) while symptomatic. This means they went to work while feeling ill. It is imperative that people stay home with even the mildest symptoms. If in doubt – do the best thing to protect public health and stay home.
Provincial COVID-19 Testing
Available with daily dashboard posting at Saskatchewan.ca/covid19
Presenter
Presentation Notes
Testing has also trended upward throughout the pandemic event.
Indicator Summary (Nov 11 -17, 2020) 7 Day Average (Nov 11-17)
ZoneDaily new case
rate per 100,000Daily active case rate per 100,000
Daily test positivity (%)
Far North West 27.7 313.8 6.6
Far North Central 214.7 1,207.5 12.3Far North East 35.4 266.3 14.1North West 15.6 178.1 6.2North Central 16.9 243.3 7.2North East 22.0 189.6 9.0Saskatoon 17.5 160.5 8.9Central West 8.0 42.8 4.3Central East 3.9 77.2 2.3Regina 8.7 122.4 4.2South West 13.0 73.7 8.0South Central 12.9 84.3 5.5South East 9.6 80.1 4.9Saskatchewan 14.5 146.9 6.7Note: 38 active cases have pending residence locations
Presenter
Presentation Notes
As with all statistics, smaller populations means that a handful of new cases may dramatically increase the rates per 100,000.
Effective Reproduction Numbers
COVID-19 Modelling Update Scenarios
16
Presenter
Presentation Notes
Dynamic Modelling Initiative led by Dr. Nate Osgood with the Computational Epidemiology & Public Health Informatics Lab (CEPHIL), University of Saskatchewan and Dr. Jenny Basran (Senior Medical Information Officer, SHA) Models: Agent-Based Model: Used for capacity planning (6 month projections) -- Dr. Kurt Kreuger Particle Filtered Compartmental Model (14-day forecasts)
Why model?• Modelling provides an understanding of where we may be heading
on COVID-19 transmission and what needs to be done to reduce the negative impact on individuals and the health system.
• Six month projections and 14 day forecasts allow us to plan and prepare our hospitals, care settings and human resources to meet the needs of future surges.
• Following the implementation of a measure, it takes 7-10 days to see changes in cases, then another week to see alteration in hospitalizations, and yet another week after that for ICU admissions to change (four weeks from behaviour change).
• The speed at which changes occur depends upon the intervention implemented, the size of the intervention, and the degree of uptake by the population.
Pandemic Wave Scenarios
Presenter
Presentation Notes
#1: What was expected – pandemic is about flattening the curve, managing the case load so as not to overwhelm the health care system #2: What we are seeing #3: Desired state – we need to reduce the curve and manage at lower case rates until there is an effective vaccine available broadly enough to impact transmission rates
Modelling Scenarios
Presenter
Presentation Notes
This modeling was used when examining the measures that have been implemented this past week. They demonstrate that we – as residents of the province -- can create significant change in our trends with adherence to best practices. � 60% mask use means 100% compliance by 60% of the population. In all the scenarios: They are built upon the measures that are were in place at the start of the week – not yet incorporating the province-wide mandatory masking requirement. Built with schools remaining open, and some mixing at home, work, hospitals and LTC homes as a realistic scenario. Mixing in restaurants, friends’ homes and grocery stores / malls had additional variables incorporated including size, duration, and mixing rate. There is also a maximum capacity for contact tracing incorporated into the model at 300 new contacts per day, after which time, delays are anticipated with added negative impact.
Applying the Scenarios
Presenter
Presentation Notes
These are the potential outcomes under those modelling scenarios, showing the development from what we had used in forecasts in spring modelling to what could occur now. Under the optimistic scenario, five hospital admissions a day is still high, but even in the next best situation, 261 daily admissions is not sustainable. To put these deaths in perspective, 4,700 deaths is almost 50% of the total number of deaths per year from all causes in Saskatchewan prior to the pandemic.
Modelling Compared to Actual Situation
Presenter
Presentation Notes
Remember – the lines and the ranges represented here are the scenarios – the arrows and asterisks indicate where the actuals are. The seven day average from November 11 is on par with the optimal scenario, but the difference between November 11 and the 14 demonstrates how quickly case counts can change a curve. We want to achieve the green curve. This is basic on the optimistic scenario which did not include the measures announced this past week. The modelling does not indicate that we need dramatic action to achieve it: 60% mask use 100% of the time, reduction in private gatherings, reducing bar/restaurant use and working from home as much as possible. But this is also NOT an invitation to think that we’re accomplished this scenario with our measures – this is forecast and not a definitive. We need all residents to follow the guidance at all times. We will continue to look at additional measures in the coming days.
14 Day Forecast of Lab Confirmed Cases(to Nov 29, 2020)
Presenter
Presentation Notes
NOTE: THESE DO NOT INCLUDE THE MEASURES IMPLEMENTED. The shaded areas represent the credibility intervals, which further widens out the forecast. This reflects individuals who are being diagnosed that day, and does not reflect when they were infected. This needs to be interpreted with caution because the model is not able to anticipate such variables as why people go for testing or testing accessibility.
14 Day Forecast of Acute Hospital Admissions (to Nov 29, 2020)
Presenter
Presentation Notes
NOTE: THIS DOES NOT INCLUDE MEASURES TAKEN TO DATE. The impact of public health measures implemented on last Friday and earlier this week will not impact on the number of hospital admissions for another two weeks. Based on how well actual hospital admissions have closely tracked to the projected hospital admissions, our hospital admissions will likely continue to increase exponentially as seen in this forecast.
Modelling Scenarios
Presenter
Presentation Notes
Reminder that this is our roadmap, guiding our collective action which can have a real impact on provincial cases. The implementation of masking orders should achieve a higher compliance rate than what we have here, but as has been repeated: mask wearing must be accompanied by physical distancing, staying home if you are sick, and all the other layers of protection.
Public Health Action: Measures Make the Difference
25
The actions taken by the people of Saskatchewan collectively will determine the outcomes.
Simple measures reduce the risk of infection:• Mask whenever you are away from your home• Physically distance• Reduce close contacts and non-essential travel• Outdoors is better than indoors• Washing hand/sanitizing often• Abide by the public health guidelines in every
public space
Everyone plays their part.
Presenter
Presentation Notes
Fundamental goal is to prevent serious illness and death and ensure that we have the health care capacity to care for COVID and non-COVID patients. Canada has benefitted immensely from our publicly-funded health care system; we inherently believe that it will be there for us but we are already testing the limits. The SHA is requiring surge capacity measures. If we want the health care system to be there for us, we need to commit to these safety measures for them.
